SmartLite: A DBMS-based Serving System for DNN Inference in Resource-constrained Environments
Summary: SmartLite is a lightweight DBMS that stores DNN parameters/structure as tables and implements neural operators, turning tensor ops into DB value-lookups. Uses binarized quantization and pruning to run multiple models on constrained edge devices, achieving 98% memory reduction and ~134% speedup vs TorchServe. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
BibTeX Citation
@article{lin_vldb24,
title = {{SmartLite: A DBMS-based Serving System for DNN Inference in Resource-constrained Environments}},
author = {Lin, Qiuru and Wu, Sai and Zhao, Junbo and Dai, Jian and Shi, Meng and Chen, Gang and Li, Feifei},
journal = {PVLDB},
series = {{VLDB} '24},
volume = {17},
number = {3},
pages = {278--291},
doi = {10.14778/3632093.3632095},
url = {https://doi.org/10.14778/3632093.3632095},
year = {2024}
}Incoming Citations (Sorted by Pagerank)
Showing 7 of 7 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 6,614 | Mitigating the Impedance Mismatch between Prediction Query Execution and Database Engine | 2025 | SIGMOD | 5.8216658e-05 |
| 8,908 | Privacy and Accuracy-Aware AI/ML Model Deduplication | 2025 | SIGMOD | 5.3483178e-05 |
| 9,542 | Database as Runtime: Compiling LLMs to SQL for In-database Model Serving | 2025 | SIGMOD | 5.2528121e-05 |
| 10,230 | Eliminating Redundant Feature Tests in Decision Tree and Random Forest Inference on SQL Predicates | 2026 | SIGMOD | 5.093636e-05 |
| 10,232 | EncoderForge: Generating Efficient SQL for Encoders in Machine Learning Inference Pipelines | 2026 | SIGMOD | 5.093636e-05 |
| 10,466 | InferF: Declarative Factorization of AI/ML Inferences over Joins | 2026 | SIGMOD | 5.093636e-05 |
| 10,750 | Alsatian: Optimizing Model Search for Deep Transfer Learning | 2025 | SIGMOD | 5.093636e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 6 of 6 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 106 | The MADlib Analytics Library or MAD Skills, the SQL | 2012 | VLDB | 0.00033539462 |
| 870 | BitWeaving: Fast Scans for Main Memory Data Processing | 2013 | SIGMOD | 0.0001350293 |
| 2,347 | Vertica-ML: Distributed Machine Learning in Vertica Database | 2020 | SIGMOD | 8.7157552e-05 |
| 2,786 | DB4ML – An In-Memory Database Kernel with Machine Learning Support | 2020 | SIGMOD | 8.1207221e-05 |
| 4,114 | Optimizing Machine Learning Inference Queries with Correlative Proxy Models | 2022 | VLDB | 6.8941194e-05 |
| 5,508 | LMFAO: An Engine for Batches of Group-By Aggregates | 2020 | VLDB | 6.1922654e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 13,432 | Using Deep Learning Models to Replace Large Materialized Views in Relational Database | 2021 | CIDR |
| 2 | 4,600 | LightNE: A Lightweight Graph Processing System for Network Embedding | 2021 | SIGMOD |
| 3 | 6,061 | VergeDB: A Database for IoT Analytics on Edge Devices | 2021 | CIDR |
| 4 | 9,566 | Declarative Data Serving: The Future of Machine Learning Inference on the Edge | 2021 | VLDB |
| 5 | 10,410 | TranSQL+: Serving Large Language Models with SQL on Low-Resource Hardware | 2026 | SIGMOD |
| 6 | 3,253 | Jointly Optimizing Preprocessing and Inference for DNN-based Visual Analytics | 2021 | VLDB |
| 7 | 10,386 | NeurStore: Efficient In-database Deep Learning Model Management System | 2026 | SIGMOD |
| 8 | 7,369 | SmartBench: A Benchmark For Data Management In Smart Spaces | 2020 | VLDB |
| 9 | 10,041 | DBMS Fitting: Why should we learn what we already know? | 2020 | CIDR |
| 10 | 6,708 | Serving Deep Learning Models with Deduplication from Relational Databases | 2022 | VLDB |